02:23:48 <Kronovestan> I'm getting a lot of "failed out" transactions however they **ARE** in the txpool on xmrchain.net at some time (5 mins about). Sometimes instantly, but very very rare. Anyone else notice this?
02:25:15 <selsta> Kronovestan: are you using your own node?
02:25:48 <Kronovestan> @selsta yes my own node.
02:25:56 <Kronovestan> This happens with remote nodes as well.
02:27:23 <selsta> Do they stay as failed?
02:27:30 <Kronovestan> I am using a ledger with CLI and have tried both a Ledger nano X and S and both get the same result. I can try cakewallet on my own node and see.
02:27:38 <Kronovestan> they CONFIRM eventually.
02:27:44 <Kronovestan> *always*
02:28:11 <Kronovestan> xmrchain.net picks them up in txpool but not after MY wallet shows failed out
02:28:49 <Kronovestan> I've tested this with minko.to as well and if you know that game the ball will drop when they see the TX.. same as xmrchain.net .. .both will see the TX at the same time.
02:28:51 <selsta> Are you connecting to a restricted remote node?
02:29:06 <Kronovestan> local node running on my own pc
02:29:46 <Kronovestan> I am running the daemon with "monerod --restricted-rpc"
02:30:26 <selsta> yea that explains it, it is Dandelion++ and restricted rpc related
02:30:50 <Kronovestan> I'm afraid of running it without that flag
02:31:27 <Kronovestan> can I do monerod --rpc-login user:pass
02:31:34 <Kronovestan> would that be safe?
02:31:49 <selsta> should work
02:33:16 <selsta> vtnerd explained the issue you are describing once, apparently it is not trivial to fix
02:35:27 <Kronovestan> ok pending out
02:35:56 <Kronovestan> failed
02:36:48 <Kronovestan> Guess I'll just have to close my ports up and just run the daemon local without the restricted flag till this is fixed.
08:16:04 <gyt0> my logs are flooded by lines: ERROR txpool src/cryptonote_core/tx_pool.cpp:139 failed to find tx meta
08:16:19 <gyt0> how to avoid it?
12:11:34 <binaryFate> binaries for CLI 0.17.0.1 are out on getmonero.org. Thanks to TheCharlatan, hyc, iDunk and scoobybejesus for the reproducible builds gitian PRs :) And selsta obviously!
12:13:44 <moneromooo> \o/ Thanks all :)
13:32:11 <Kronovestan> The binaries on getmonero.org are still 0.17.0.0 but it says there is a new version?
13:34:14 <Kronovestan> getmonero.org is still giving out Version: 0.17.0.0 for Windows 64-bit
13:35:17 <binaryFate> I get 0.17.0.1
13:35:55 <binaryFate> Your browser probably cached it. Try with incognito window or different browser?
13:37:25 <Kronovestan> That works... incognito window.
13:37:39 <Kronovestan> *glares at firefox*
13:38:08 <moneromooo> Firefox reports Kronovestan to the mozilla mothership
13:38:08 <Kronovestan> I wanted to go revenge bet more at minko.to lol
13:38:30 <binaryFate> solid plan
13:40:45 <Kronovestan> it's funny binary I can bet min for tons of times and be fine but then big one .. boom loss.. so odd
13:44:13 <binaryFate> thanks provable fairness, you can check all that tomorrow
13:45:18 <Kronovestan> ok
13:48:41 <Kronovestan> When you send a TX it seems to take a long time to go though the network now ... for other sites to see the TX takes up to 5 mins. Is this new? It used to be instant.
13:52:14 <moneromooo> It is new. It's due to Dandelion, which sends txes to just one (or two ?) or your peers, which in turn do the same. It used to be sent to all peers at once.
13:52:44 <moneromooo> The intent is to make it harder for someone to get an overall view of the network.
13:53:13 <moneromooo> But if a node along the stem phase drops txes on the floor, it takes a wihle for a timeout to occur and txes to be broadcasted.
13:53:55 <binaryFate> could the timeout be shorten maybe?
13:54:21 <sech1> if you use something like xmr.to you can just connect directly to their node (node.xmr.to) to send tx instantly
13:55:14 <moneromooo> IIRC the timeout is not arbitrary (well, not totally arbitrary).
14:02:58 <selsta> It will be interesting to see if it gets better or worse with v0.17 when all legit nodes use Dandelion.
14:30:28 <Inge-> hm. I haven't seen 5 minutes - more like 30-45 seconds vs 1-5 seconds before
14:40:56 <Guest_91> Hi guys
14:41:13 <Guest_91> Is there exists guide to setup dev env for monero?
14:41:48 <Guest_91> My problem is errors when i try to run python tests
14:41:57 <Guest_91> Got this error: "ModuleNotFoundError: No module named 'framework'"
14:43:25 <Guest_91> How to fix this?
14:44:11 <selsta> Which OS?
14:46:42 <Guest_91> MacOS
14:48:07 <Kronovestan> what python is your env using?
14:48:22 <Guest_91> 3.8
14:49:18 <Kronovestan> You have Xcode installed?
14:53:54 <selsta> you have to install requests with pip
14:54:14 <selsta> but i had some annoying python2 / python3 issues on mac, don’t remember currently how I solved it
15:09:09 <Guest_91> xcode is installed
15:09:35 <Kronovestan> Oh damn. From the time of reply I was hoping that was the issue and it was fixed.
15:13:02 <sech1> moneromooo transactions propagate much slower now with 0.17. I waited for a couple of minutes before my tx appeared on xmrpool.net
15:13:08 <sech1> I was on 0.15 before
15:24:17 <selsta> It's due to Dandelion++ and most likely hostile nodes black holing transactions.
15:28:21 <selsta> It's only slow sometimes, depending on which peer gets the tx propagated.
15:28:23 <sech1> what's the timeout for resending tx? Does resending go through another peer?
15:29:48 <selsta> average is 173 seconds
15:29:57 <Inge-> selsta: how do you time that?
15:30:08 <selsta> It's in the code
15:31:06 <selsta> sech1: resending is without D++ and it goes to all peers AFAIK
15:31:27 <sech1> hmm, so that opens a sybil attack vector
15:32:02 <sech1> I had 2 transactions today and both had to be resent
15:32:11 <selsta> manually?
15:32:20 <sech1> first one manually, it didn't propagate at all
15:32:26 <sech1> second one went through after a few minutes
15:32:44 <sech1> I use my own node
15:33:44 <selsta> see 4.4 https://arxiv.org/pdf/1805.11060.pdf
15:35:21 <sech1> all right, I'll keep an eye on propagation when I do next transactions. This can only be tested on mainnet.
15:36:11 <moneromooo> I guess we need to find a good way to get rid of those nodes. They're obvious atm, but if we start weeding them out, they'll start hiding.
15:49:53 <Kronovestan> I've been playing with sending TX and they are taking up to 5 mins some times to show on xmrchain.net or my other node on another ISP. Some times it is instant. I dono if it's Dandelion++ or not. How can we test?
15:51:56 <moneromooo> You could disable dandelion. This would be... setting a flag to 1 or 0 in the transactions packet. Let me see..
15:57:14 <sarang> Research report is available: https://www.reddit.com/r/Monero/comments/j2oncd/september_monthly_report_from_sarang_noether/
16:52:36 <Kronovestan> @moneromooo, I don't see how to set the dandelion
17:26:19 <moneromooo> I PMed you the code.
19:40:38 <dsc_> on what unix epoch did monero blockchain start
19:40:45 <dsc_> 18 apr 2014
19:41:31 <dsc_> 1397818193
19:41:33 <dsc_> hmmz
19:41:36 <moneromooo> Epoch ?
19:41:46 <dsc_> unix timestamp*
19:41:47 <moneromooo> 1/1/1970, if you really mean epoch.
19:42:03 <moneromooo> Ah. That date looks plausible.
19:42:16 <dsc_> Ill go with `1397818193`
19:42:16 <moneromooo> The genesis block ts might be wrong though.
19:43:41 <UkoeHB_> yeah genesis doesn't have a timestamp, but block 1 does
19:43:56 <azy> thats my birthday!
19:52:34 <binaryFate> genesis block wasn't mined so block 1 better representative of actual launch